Showing papers on "Robustness (computer science) published in 2022"
TL;DR: This paper provides a review on the use of machine learning techniques in the design of different elements of meta-heuristics for different purposes including algorithm selection, fitness evaluation, initialization, evolution, parameter setting, and cooperation.
106 citations
TL;DR: In this paper, an unsupervised domain-share convolutional neural network (CNN) is proposed for efficient fault transfer diagnosis of machines from steady speeds to time-varying speeds.
93 citations
TL;DR: A new GI evaluation frame is built, including the definition of new indexes based on GI, and enhancing signal processing methods via GI, such as spectrum kurtosis, decomposition methods, and multi-objective optimization algorithms are designed.
70 citations
TL;DR: A quantitative tuning rule for the time-delayed ADRC (TD-ADRC) structure based on the typical first order plus time delay (FOPTD) model is proposed, revealing Relative delay margin is revealed as a critical robustness metric among others.
Abstract: Active disturbance rejection controller (ADRC) has achieved soaring success in motion controls featured by rapid dynamics. However, it turns obstreperous to implement it in the power plant process with considerable time-delay, largely because of the tuning difficulty. To this end, this article proposes a quantitative tuning rule for the time-delayed ADRC (TD-ADRC) structure based on the typical first order plus time delay (FOPTD) model. By compensating the FOPTD process as an integrator plus time delay in low frequencies, the gain parameter of TD-ADRC can be related to a scaled time constant which shapes the closed-loop tracking performance. Bandwidth parameter of extended state observer is scaled as a dimensionless parameter. A sufficient stability condition of TD-ADRC is theoretically derived in terms of the scaled parameter pair, the range of which falls within the practical interest. Relative delay margin is revealed as a critical robustness metric among others, a default pair of scaled parameter setting is recommended as well as an explicit retuning guideline according to the user's preference for performance or robustness. Simulation and laboratory water tank experiment validate the tuning efficacy and a coal mill temperature control test depicts a promising prospective of the proposed method in process control practice.
64 citations
01 Jan 2022
TL;DR: An optimization scheme based on the Alternating Direction Method of Multipliers that minimizes not only the training loss of an NN but also its Lipschitz constant resulting in a semidefinite programming based training procedure that promotes robustness.
Abstract: Due to their susceptibility to adversarial perturbations, neural networks (NNs) are hardly used in safety-critical applications. One measure of robustness to such perturbations in the input is the Lipschitz constant of the input-output map defined by an NN. In this letter, we propose a framework to train multi-layer NNs while at the same time encouraging robustness by keeping their Lipschitz constant small, thus addressing the robustness issue. More specifically, we design an optimization scheme based on the Alternating Direction Method of Multipliers that minimizes not only the training loss of an NN but also its Lipschitz constant resulting in a semidefinite programming based training procedure that promotes robustness. We design two versions of this training procedure. The first one includes a regularizer that penalizes an accurate upper bound on the Lipschitz constant. The second one allows to enforce a desired Lipschitz bound on the NN at all times during training. Finally, we provide two examples to show that the proposed framework successfully increases the robustness of NNs.
59 citations
TL;DR: A novel framework using Bidirectional Gated Recurrent Unit (Bi-GRU) and Sparrow Search Algorithm (SSA) that improves the accuracy of oil rate forecasting and is compared with traditional decline curve analysis, conventional time series methods and one-way recurrent neural networks.
55 citations
TL;DR: A distributed robust economic dispatch strategy is presented to achieve the energy management of IESs in the presence of misbehaving units, which consists of a neighbor-observe-based detection process and a reputation-based isolation process that possesses strong robustness against various colluding and noncolluding attacks.
Abstract: Distributed algorithms are increasingly being used to solve the economic dispatch problem of integrated energy systems (IESs) because of their high flexibility and strong robustness, but those algorithms also bring more risk of cyber-attacks in IESs. To solve this problem, this article investigates the distributed robust economic dispatch problem of IESs under cyber-attacks. First, as the first line of defense against attacks, a privacy-preserving protocol is designed for covering up some vital information used for economic dispatch of IESs. On this basis, a distributed robust economic dispatch strategy is presented to achieve the energy management of IESs in the presence of misbehaving units, which consists of a neighbor-observe-based detection process and a reputation-based isolation process. The proposed strategy is implemented in a fully distributed fashion and possesses strong robustness against various colluding and noncolluding attacks. In addition, the strategy can not only ensure the reliability of information transmission among energy units, but also solve the problem of incorrect measurement of distributed local load data caused by cyber-attacks. Finally, the effectiveness of the proposed strategy is illustrated by simulation cases on a 39-bus 32-node power–heat IES.
54 citations
TL;DR: In this article, a novel fusion-based SOH estimator is proposed, which combines an open circuit voltage (OCV) model and the incremental capacity analysis, and the extracted model FOI and IC morphological FOIs are further fused for SOH estimation through an artificial neural network.
Abstract: The state of health (SOH) is a vital parameter enabling the reliability and life diagnostic of lithium-ion batteries. A novel fusion-based SOH estimator is proposed in this study, which combines an open circuit voltage (OCV) model and the incremental capacity analysis. Specifically, a novel OCV model is developed to extract the OCV curve and the associated features-of-interest (FOIs) from the measured terminal voltage during constant-current charge. With the determined OCV model, the disturbance-free incremental capacity (IC) curves can be derived, which enables the extraction of a set of IC morphological FOIs. The extracted model FOI and IC morphological FOIs are further fused for SOH estimation through an artificial neural network. Long-term degradation data obtained from different battery chemistries are used for validation. Results suggest that the proposed fusion-based method manifests itself with high estimation accuracy and high robustness.
51 citations
TL;DR: It is proved that the proposed AFNTSM can guarantee finite-time convergence and zero tracking error for the quadrotor attitude control and can achieve faster convergence and stronger robustness in line with theoretical analysis.
Abstract: As one type of unmanned aerial vehicles, the quadrotor typically suffers from payload variations, system uncertainties, and environmental wind disturbances, which significantly deteriorate its attitude control performance. To provide high-speed, accurate, and robust attitude tracking performance for the quadrotor, an adaptive fast nonsingular terminal sliding mode (AFNTSM) controller is proposed in this article. The proposed AFNTSM controller combines the advantages of fast nonsingular terminal sliding mode (FNTSM), integral sliding mode, and adaptive estimation techniques, which are effective to achieve the desired tracking performance and suppress control signal chattering. Furthermore, unlike conventional methods, the adaptive estimation removes the requirements for the upper bound information of the disturbances. It is proved that the proposed AFNTSM can guarantee finite-time convergence and zero tracking error for the quadrotor attitude control. Finally, comparative study with the FNTSM control only and conventional sliding mode control is conducted through experiments and the results demonstrate that the proposed AFNTSM can achieve faster convergence and stronger robustness in line with theoretical analysis.
49 citations
TL;DR: In this article, a multi-feature-based multi-model fusion method is proposed for the SOH estimation of lithium-ion batteries, where the key factors of the battery aging process are analyzed from multiple sources such as voltage, temperature, and incremental capacity curves.
44 citations
TL;DR: A federated learning scheme combined with the adaptive gradient descent strategy and differential privacy mechanism is proposed, which is suitable for multi-party collaborative modeling scenarios and shows robustness to different super-parameter settings.
TL;DR: The numerical study and the experimental data both demonstrate that the proposed automated tuning method is efficient in terms of required tuning iterations, robust to disturbances, and results in improved tracking.
Abstract: In this article, we propose a performance-based autotuning method for cascade control systems, where the parameters of a linear axis drive motion controller from two control loops are tuned jointly. Using Bayesian optimization as all parameters are tuned simultaneously, the method is guaranteed to converge asymptotically to the global optimum of the cost. The data-efficiency and performance of the method are studied numerically for several training configurations and compared numerically to those achieved with classical tuning methods and to the exhaustive evaluation of the cost. On the real system, the tracking performance and robustness against disturbances are compared experimentally to nominal tuning. The numerical study and the experimental data both demonstrate that the proposed automated tuning method is efficient in terms of required tuning iterations, robust to disturbances, and results in improved tracking.
TL;DR: Experimental results consistently demonstrate the proposed pre-compensation scheme can achieve the tracking accuracy comparable to iterative learning, while maintaining the robustness to trajectory changes and uncertain disturbances without re-offline iteration.
Abstract: In this article, to guarantee the good tracking performance of the precision motion system for various tracking tasks, an online iterative learning compensation method is proposed for closed-loop motion control systems. The prediction model is based on the closed-loop model of the linear second-order system with a proportional-integral-derivative controller, and an estimation term is added to deal with the influence of slow-varying uncertain disturbances. On the basis of the accurate state prediction, the dynamical feedforward compensation can be obtained, which suppresses the tracking error caused by the dynamical lag. Furthermore, in order to simultaneously compensate the errors caused by nonlinear factors such as uncertain disturbances and to guarantee the smoothness of the compensated trajectory, the optimal compensation gain is determined through online iterative calculation. The online iterative approach is similar to iterative learning control, but does not require several offline iterations of a repeating trajectory. Comparative experiments are carried out on an industrial motion stage. Various experimental results consistently demonstrate that the proposed compensation scheme can achieve the tracking accuracy comparable to iterative learning, while maintaining the robustness to trajectory changes and uncertain disturbances without reoffline iteration.
TL;DR: The extended deep sequence-to-sequence long short-term memory regression (STSR-LSTM) model is used for time-series wind power forecasting to overcome challenges for accurate forecasting decisions and achieve higher forecast accuracy.
TL;DR: This work proposes a black-box adversarial attack for crafting adversarial samples to test the robustness of clustering algorithms, formulated as a constrained minimization program, general in its structure and customizable by the attacker according to her capability constraints.
TL;DR: Although the lifespan of LIBs is estimated using the training set in the 5 % SOH range, the estimation errors of the proposed framework are less than 2.5 % in all test sets, ensuring its potential applicability in practical implementations of onboard battery management systems.
TL;DR: This work proposes a selective fixed-filter ANC method based on a simplified two-dimensional convolution neural network (2D CNN), which is implemented on a co-processor (e.g., in a mobile phone), to derive the most suitable control filter for different noise types.
TL;DR: In this paper, a small sample fault diagnosis method is proposed based on dual path convolution with attention mechanism (DCA) and bidirectional gated recurrent unit (BiGRU), whose performance can be effectively mined by the latest regularization training strategies.
TL;DR: In this paper, a Gaussian parametric level set method coupled with regularized Landweber is proposed to obtain the posterior topological structure of the turbulent reaction flow, which can compensate for the lack of a priori knowledge of the tomographic inverse problem.
TL;DR: Wang et al. as mentioned in this paper put forward an optimized structure-adaptive grey model by theoretically providing the generalized time response function and accurately modifying the background value based on Simpson's rule to predict nuclear energy consumption in China and America.
TL;DR: The KWAY project as discussed by the authors investigated the economic sustainability of up-front NGS technologies adoption in the analysis of clinically relevant molecular alterations in NSCLC patients, and highlighted that the adoption of NGS allows to save personnel time dedicated to testing activities and to reduce the overall cost of testing per patient.
Abstract: Aims The KWAY project aims to investigate the economic sustainability of the up-front NGS technologies adoption in the analysis of clinically relevant molecular alterations in NSCLC patients. Methods The diagnostic workflow and the related sustained costs of five Italian referral centers were assessed in four different evolving scenarios were analyzed. For each scenario, two alternative testing strategies were evaluated: the Maximized Standard strategy and the Maximized NGS strategy. Results For each center, the robustness of obtained results was verified through a deterministic sensitivity analysis, observing the variation of total costs based on a variation of ±20 % of the input parameters and ensuring that results would present a consistent behavior compared to the original ones. Conclusions our project, highlighted that the adoption of NGS allows to save personnel time dedicated to testing activities and to reduce the overall cost of testing per patient.
TL;DR: A kernel MSE loss function to evaluate the ubiquitous nonlinearity of deep learning errors in the reproducing kernel Hilbert space is proposed and the results imply that developing a loss function with kernel skills is a new way to get better results.
TL;DR: A cloud-edge collaboration strategy that integrates multi-model adaptation and machine learning is proposed for battery capacity prediction in lifespan and multiple individual algorithms are promoted relying on online data from battery management system and induced ordered weighted average operator is introduced for joint capacity estimation.
TL;DR: A resilience assessment framework of the urban road networks, including the resilience performance index, the robustness index, and the recovery index is proposed, and results show that the intersection-based attack has the most significant impact on resilience, and resilience is positively correlated with the node degree of the attacked intersection.
Abstract: Urban space for new transportation facilities cannot meet the increasing traffic demand. Afterward, scholars gradually increased attention to the resilience evaluation of urban road networks. Therefore, we proposed a resilience assessment framework of the urban road networks, including the resilience performance index, the robustness index, and the recovery index. Then we simulated the cascading failure based on a nonlinear load-capacity model with two capacity control parameters: α and β . Results show that the intersection-based attack has the most significant impact on resilience, and resilience is positively correlated with the node degree of the attacked intersection. The increase of α and β could enhance the resilience, and the urban road network achieves the best resilience performance when α = 0 . 3 , β = 0 . 5 . Compared with the deliberate attack strategy, the resilience performance under the random attack strategy is more robust. This research can provide the foundation for optimizing urban road networks and multi-mode urban public transit networks.
SIDI1
Abstract: In this work, we first present a modified version of the traditional logistic chaotic map. The proposed version contains an additional parameter that is used to increase the security level of the proposed digital image copyright protection scheme. The latter merges two methods of image copyright protection, namely the image zero-watermarking and image encryption, which provides a high level of security when communicating images via the Internet. Next, we discuss the influence of geometric attacks on the efficiency of the proposed scheme, and then we introduce an efficient solution that can resist such attacks. The proposed solution involves the use of Sine Cosine Algorithm (SCA) with an appropriate algorithm suitable for the correction of geometric attacks (image translation, orientation and its combination) applied to the encrypted image. On the one hand, the simulation results show that the proposed scheme provides a high level of security and can resist various attacks (differential, common image processing, geometric, etc.). On the other hand, the conducted comparison in terms of robustness against geometric attacks clearly demonstrates the superiority of our scheme over recent image encryption ones.
TL;DR: The results show that the proposed system not only reduces the amount of data transmission, but also improves the precision of the reconstructed image and the robustness of the encryption system.
TL;DR: Simulation and experiment results show that the proposed RHC scheme solved by the RNN model is able to make the redundant manipulator track the given trajectory excellently, and is superior to other existing schemes and solvers in terms of high efficiency, quick-response capacity, and strong robustness.
Abstract: Redundant manipulators have been studied and applied in many fields. The trajectory tracking of redundant manipulators is an important topic to explore for applications. This article aims to develop a planning scheme for achieving the trajectory tracking of redundant manipulators, from the receding horizon control (RHC) perspective. For the nonlinear model of manipulators, the linearization operation is conducted to obtain predictive outputs through the forward kinematic equation. Subsequently, an RHC scheme, which minimizes tracking error, velocity norm, and acceleration norm, and directly considers joint limits at three levels as well as the terminal equality constraint, is constructed and further simplified as a convex quadratic programming problem. Furthermore, a recurrent neural network (RNN) model is designed for the constructed RHC scheme, with the help of the technique of converting inequality constraints into equality constraints. The proposed RHC scheme solved by the RNN model is compared with other existing planning schemes and solvers through computer simulations and experiments, without and with the sudden external interference. Simulation and experiment results show that the proposed RHC scheme solved by the RNN model is able to make the redundant manipulator track the given trajectory excellently, and is superior to other existing schemes and solvers in terms of high efficiency, quick-response capacity, and strong robustness.
TL;DR: A robust BC controller is designed for PMSG-based WECS operations during normal and grid fault conditions and the simulation results reflect on effectuality and robustness of the proposed BC controller in comparison to conventional BC control strategies.
DOI•
01 Jan 2022TL;DR: In this paper, a swarm intelligence-based algorithm, brainstorm optimization, is proposed for reducing dimensionality (feature selection) in datasets that are used for classification, which is a well-known and widely used technique in analyzing big data.
Abstract: In this work, a swarm intelligence-based algorithm, brainstorm optimization, is proposed for reducing dimensionality (feature selection) in datasets that are used for classification. Dimensionality reduction is a well-known and widely used technique in analyzing big data. Its role is to reduce the number of features in high-dimensional datasets and to keep only those that contain useful and rich information. This results in better understanding and interpretation of data, higher accuracy, and boosting the training process of machine learning method used for classification. After extracting features from the dataset, it should be decided which subset of features will be used in the training process. Since, in high-dimensional datasets many features exist, this problem is categorized as NP hard and it is necessary to employ metaheuristics for its solving. For tackling this issue, a binary hybrid brainstorm optimization metaheuristics that overcome the drawbacks of original algorithm, is proposed. For performance evaluation, 21 datasets are used. The comparative analysis is made between the proposed approach and the original brainstorm optimization algorithm, as well as with nine other metaheuristics adopted for feature selection. Experimental results prove the robustness of proposed method, since it is capable to reduce the number of features by simultaneously achieving better classification accuracy than other methods taken for comparative analysis.
TL;DR: The proposed method can be utilized for different multiterminal dc systems and is effective under different fault locations, different fault types, and high fault impedances, and does not require high sampling frequency and has good robustness against measuring noise.
Abstract: Existing nonunit protection schemes inevitably require setting, which is a serious problem in practical engineering. Faults occurred at different fault zones will result in different equivalent models, therefore, the fault zone can be determined by recognizing which equivalent model the fault fits well with. In this article, this “model recognition” idea is introduced in fault identification and a “setting-less” protection method is proposed. First, the Peterson equivalent circuits when faults occur at backward external zone, internal zone, and forward external zone are presented, respectively. Accordingly, the corresponding three fault voltage expressions are derived, which are defined as three fault modes. Then, the three fault modes are used to approximate the measured fault voltage using Levenberg-Marquardt optimal approximation method. The fault mode that best fits the measured fault voltage is recognized as the final determined fault mode, which is used for fault identification without setting threshold value. Numerous test studies carried out in Power Systems Computer Aided Design/Electromagnetic Transients including DC (PSCAD/EMTDC) and real-time digital simulator have demonstrated that the proposed method can be utilized for different multiterminal dc systems and is effective under different fault locations, different fault types, and high fault impedances. The proposed method does not require high sampling frequency and has good robustness against measuring noise.